Local invariant feature detectors: a survey
Foundations and Trends® in Computer Graphics and Vision
An efficient image-mosaicing method based on multifeature matching
Machine Vision and Applications
A comparative study of different corner detection methods
CIRA'09 Proceedings of the 8th IEEE international conference on Computational intelligence in robotics and automation
Machine learning for high-speed corner detection
ECCV'06 Proceedings of the 9th European conference on Computer Vision - Volume Part I
Schelling points on 3D surface meshes
ACM Transactions on Graphics (TOG) - SIGGRAPH 2012 Conference Proceedings
The feature detection on the homogeneous surfaces with projected pattern
ITIB'12 Proceedings of the Third international conference on Information Technologies in Biomedicine
| Hi-index | 0.00 |
Selecting salient points from two or more images for computing correspondences is a fundamental problem in image analysis. Three methods originally proposed by Harris et al. in [A combined corner and edge detector], by Noble et al. in [Descriptions of image surfaces] and by Shi et al. in [Good features to track] proved to be quite effective and robust and have been widely used by the computer vision community. The goal of this paper is to analyze these point detectors starting from the algebraic and numerical properties of the image auto-correlation matrix. To accomplish this task we will first introduce a "natural" constraint that needs to be satisfied by any point detector based on the auto-correlation matrix. Then, by casting the point detection problem in a mathematical framework based on condition theory [A condition number for point matchingwith application to registration and post-registration error estimation], we will show that under certain hypothesis the point detectors are equivalent modulo the choice of a specific matrix norm. The results presented in this paper will provide a novel unifying description for the most commonly used point detection algorithms.